The past literature on marital dissolution has documented that the children and the custodial parent often experience acute and lasting emotional, social, cognitive and financial distress. Unexpectedly, there is also evidence that the noncustodial parent (NCP) suffers at least as much in the short term, primarily because of the loss of the child. The past literature also demonstrates that a common manifestation of the NCP's distress is decreased involvement, both personal and financial (child support), with his child. Unfortunately, there is little in this past work to explain why some NCPs continue their involvement with their children and others do not. The present project proposes a four-year longitudinal study to determine the factors that predict the nature of the NCP-child relationship after separation. We developed a conceptual model based upon a social exchange theoretic perspective in which the NCP-child relationship is viewed as subject to all the usual principals of dyad-maintenance. Using the cost-benefit analysis that is thus implied we will examine the material, symbolic, and affectional rewards and costs that should predict continued involvement. As rewards increase or costs decrease we expect more involvement. From a sample of 20 recently divorced families, which will be more representative than those in past work, we will gather interview data as well as archival child-support records. The interviews will follow a structured format using pre-established measures from the literature and from our own past research. Additional measures will be developed especially for the project. Special measures will be taken to prevent or correct for attrition. Our cost-benefit analysis model will be tested using the improper linear model technique. We also propose exploratory regression analysis, to provide a wider analysis of the factors that predict continued involvement. Finally, as subsidiary analyses we analyze the predictors of custodial parent's, child's, and NCP's post-divorce adjustment.